TGF-? is released from cells primarily as a large latent complex (LLC) consisting of the TGF-? homodimer, the TGF-? propeptide dimer, and a single molecule of the latent TGF-? binding protein (LTBP), which is disulfide bonded to each of the propeptide chains. The propeptide, which remains associated with TGF-? even though the bond between propeptide and TGF-? has been cleaved, renders the TGF-? latent. TGF-? must be freed from the LLC to bind to its receptors. The LLC is believed to be the biologically significant form of the cytokine. A minor fraction of the latent TGF-? is secreted complexed to its propeptide in the absence of LTBP. This form is called the small latent complex (SLC). The functional differences between the LLC and SLC of latent TGF-?, which differ only by the presence or absence of LTBP, are unknown. To interrogate the functions of the SLC, we generated mutant mice (Tgfb1C33S/C33S and Tgfb1-/C33S) in which the cysteines in the TGF-?1 propeptide that bind to LTBP were replaced by serines. This mutant TGF-?1 propeptide cannot bind to LTBP. Thus, none of the secreted latent TGF-?1 is bound to an LTBP, but both LTBP and TGF-?1 complexed to its propeptide (SLC) are secreted normally. Tgfb1C33S/C33S mice have multi- organ inflammation, develop tumors of the stomach, colon, rectum and anus, and die by 3-4 months. This phenotype resembles that of the Tgfb1-/- mouse except that the Tgfb1C33S/C33S inflammatory phenotype is milder but the tumor phenotype stronger than the Tgfb1-/- phenotypes. Interestingly, 100% of Tgfb1-/C33S mice develop gastric adenocarcinomas by 12 weeks of age and have a stronger inflammatory response compared to Tgfb1C33S/C33S mice. These results indicate 1) that LTBP binding to the TGF-? propeptide is important for proper TGF-? function, 2) that there are mechanisms to activate the latent TGF-? propeptide complex in the absence of LTBP, and 3) that Tgfb1C33S/C33S and Tgfb1-/C33S mice provide a new model system to study the role of TGF-? in GI tumor induction. Using Tgfb1C33S/C33S and Tgfb1-/C33S mice in the experiments described in this application, we will establish what are the activators for the SLC, the contribution of inflammation to tumor production - specifically gastric adenocarcinomas, the changes in cell proliferation and apoptosis in the GI epithelium, and the biochemical interactions between the epithelium and stroma during gastric tumor development. We will use a combination of mouse genetics, cell biology, and biochemistry to answer these questions. The successful completion of the proposed experiments will provide information critical to the understanding of the control of TGF-? action. Such information may yield insights for therapeutic intervention or prevention of pathological conditions, such as fibrosis, cancer, and autoimmune disease that arise from inappropriate activation and activity of TGF-?. PUBLIC HEALTH RELEVANCE: Understanding how the activation of latent TGF-? is controlled may have impact into treating conditions, such as fibrosis, in which excess latent TGF-? activation occurs. Our earlier observation that the integrin subunit ?6 is involved in latent TGF-? activation has initiated work in several pharmaceutical companies to make inhibitors of ?v?6 mediated TGF-? formation. In addition, the mutant mice we have produced provide a novel system for the study of the rapid formation of gastric adenocarcinomas.